DeepUnifiedMom: Unified Time-series Momentum Portfolio Construction via Multi-Task Learning with Multi-Gate Mixture of Experts
By Joel Ong, Dorien Herremans
Rating
1417
Battle Count: 294
Relevance
9/10
Highly relevant for momentum-based strategies and portfolio optimization in quantitative trading
Implementation Complexity
7/10
Requires deep learning expertise and understanding of financial markets, but code is provided
Reproducibility
4/5
The paper provides detailed methodology and experimental setup. Code is available on GitHub.
About this paper
Methodology: DeepUnifiedMom. Problem types: Time Series Forecasting, Portfolio Optimization.
The interactive Everscope explorer (charts, battles, favorites) loads below.